The Cancer Modeling Section seeks to elucidate the complex molecular/genetic program governing tumor genesis and progression through the development and analysis of genetically engineered mouse models of human cancer. Our efforts in this regard are focused on two tumor types, cutaneous malignant melanoma and the pediatric malignancy rhabdomyosarcoma. Rhabdomyosarcoma, accounting for up to 10% of all pediatric neoplasms and for more than 50% of pediatric soft tissue sarcomas, is believed to arise from imbalances in skeletal muscle cell proliferation and differentiation (Merlino and Khanna, Genes & Dev. 21: 1275-9, 2007). There are two major subtypes of rhabdomyosarcoma. Embryonal rhabdomyosarcoma, the most common subtype, typically occurs in infants and young children. There is no distinct molecular event that characterizes embryonal rhabdomyosarcoma, but they tend to demonstrate severe genomic instability. Alveolar rhabdomyosarcomas are highly aggressive tumors of adolescents and young adults, with classification based on translocations involving the genes encoding the forkhead transcription factor, and either PAX3 or PAX7. Detailed molecular pathways associated with rhabdomyosarcoma had been poorly characterized, due in part to the lack of a rhabdomyosarcoma-prone animal model. We have developed the first relevant model for embryonal rhabdomyosarcoma, showing that virtually all mice harboring a hepatocyte growth factor/scatter factor transgene (which deregulates its tyrosine kinase receptor MET) and deficient in Ink4a/Arf rapidly succumbed to highly invasive rhabdomyosarcoma (Sharp et al., Nature Med. 8: 1276-80, 2002). Highly comparable molecular lesions have also been described for human rhabdomyosarcoma. These data provide genetic evidence that MET and INK4a/ARF pathways represent critical and synergistic targets in rhabdomyosarcoma pathogenesis, and suggest a rational therapeutic combination to combat this pediatric sarcoma. Our more recent in vivo analyses of the pathways downstream of INK4a/ARF revealed that it was the ARF-MDM2-p53 pathway, and not the INK4a-CDK4-pRb pathway that was the key regulator of this disease (Ha et al., Proc. Natl. Acad. Sci. 104: 10968-73, 2007). A panel of highly and poorly metastatic cell lines was subsequently established from the many rhabdomyosarcoma tumors arising in our novel mouse model, and used in concert with microarray-based expression profiling to identify a set of genes associated with enhanced metastatic behavior. Multiple functional in vivo studies confirmed that the cytoskeletal linker EZRIN and the homeodomain-containing transcription factor SIX1 both have essential roles in determining the metastatic fate of rhabdomyosarcoma cells (Yu et al., Nature Med. 10: 175-81, 2004). SIX1 was especially intriguing, as it is known to be required for skeletal muscle development. Notably, EZRIN and SIX1 expression levels were also both enhanced in human rhabdomyosarcoma tissue samples, significantly correlating with clinical stage. Remarkably, subsequent molecular analyses showed that the EZRIN gene was in fact a direct transcriptional target of SIX1, and indispensable for SIX1-mediated rhabdomyosarcoma metastasis (Yu et al., Cancer Res. 66: 1982-9, 2006). Recently, we have found that SIX1 regulates EZRIN expression, at least in part, through epigenetic modification of the chromatin around the EZRIN gene locus, including regulating the states of methylation and acetylation within the histone tails. Studies are currently underway to more fully elaborate mechanisms by which the EZRIN gene is regulated, and also to determine the efficacy of preclinical drug studies that use histone deacetylase inhibitors or demethylating agents in concert with EZRIN knockdown to treat rhabdomyosarcoma. EZRIN appears to represent a very promising therapeutic target for patients with advanced stage rhabdomyosarcoma.